Download LaPolla, Randy J. 2009. “Chinese as a Topic

LaPolla, Randy J. 2009. “Chinese as a Topic-Comment (Not Topic-Prominent and Not SVO)
Language”. In Studies of Chinese Linguistics: Functional Approaches, ed. by Janet Xing, 922. Hong Kong: Hong Kong University Press.
1
Chinese as a topic-comment (not topicprominent and not SVO) language
Randy J. LaPolla*
Many linguists in China and the West have talked about Chinese as a topic-comment
language, that is, a language in which the structure of the clause takes the form of
a topic, about which something is to be said, and a comment, which is what is said
about the topic, rather than being a language with a subject-predicate structure like
that of English. Y. R. Chao (1968), for example, said that all Chinese clauses have
topic-comment structure and there are no exceptions. The fact that some of these
linguists, e.g. Y. R. Chao (1968) and Lü Shuxiang (1979: 70–73), used the terms
“subject” when writing in English, or zhǔyǔ when writing in Chinese, for what I
will consistently call “topic” has confused some later linguists who did not pay
attention to the definition of these terms given in those works, and so assumed that
the “subject” referred to was equivalent to what we call “subject” in English. In fact,
those authors were simply talking about topic, not grammaticalized subject. This
confusion has also led some linguists to refer to Chinese as a Subject-Verb-Object
(SVO) language,1 that is, one where word order is determined by or determines the
grammatical relations “subject” and “object.” One line of description that differed
from both of these views was that of Li and Thompson (1976, 1981: 15–20), who
argued that “subject” in Chinese is not equivalent to topic, as, unlike topic, it must
have “a direct semantic relationship with the verb as the one that performs the
action or exists in the state named by the verb” (1981: 15), but is also not exactly
equivalent to “subject” in English, as in Chinese “ ‘subject’ is not a structurally
definable notion” (1981: 19). In this view, the clause has a “subject” (defined on
semantic rather than grammatical grounds), but there is often a topic (defined as
some topical element other than “the one that performs the action or exists in the
*
I would like to thank Professor Xu Liejiong, Professor Janet Xing, and an anonymous
reviewer for helpful comments on a draft of this chapter.
stud of Chinese Book-3.indd 9
2008/11/17 12:51:04 PM
10 Randy J. LaPolla
state named by the verb”) that precedes the “subject” as well. Based on this, they
developed the idea of dividing languages into two types: “subject-prominent” (e.g.
English) and “topic-prominent” (e.g. Chinese). Both types have both “subject”
and “topic,” but the prominence of “subject” vs. “topic” differs in the two types.
This view has been very influential in discussions of Chinese structure. As can be
seen from the quote above, what Li and Thompson called “subject” is in fact the
semantic role of actor, and say that there is no grammatically definable subject. In
saying that there is no grammatically definable subject, they are agreeing with Chao
and Lü, but differ from Chao and Lü in defining subject as agent, whereas Chao and
Lü each explicitly said their concepts of subject are not related to semantic role, but
are simply topics.
As can be seen from this brief discussion, although all three approaches use the
terms “subject” and “topic,” the meaning of those terms differs in each approach.
For a number of years I have been arguing that Chinese is unlike English in
terms of the organization of clause structure, as it has not grammaticalized the kind
of restricted neutralizations of semantic roles in certain constructions which aid in
referent role identification and tracking that characterize what we call “subject”
in English (LaPolla 1988a, 1988b, 1990, 1993; Van Valin and LaPolla 1997, Ch.
6; LaPolla and Poa 2006),2 essentially supporting the view of Y. R. Chao and Lü
Shuxiang. I have argued that the structure of the clause is instead based on the
pragmatic relations of topical vs. focal material, with topical material preceding the
verb, and focal material following the verb (LaPolla 1995; LaPolla and Poa 2005,
2006).3 In this view, topic is not something unusual and separate from “the one that
performs the action or exists in the state named by the verb”; it is not defined in
any semantic terms; it is simply what the comment is about. There has been much
discussion in the literature about the nature of topics and their different types (see, for
example, Chafe 1976; Lambrecht 1994). In this chapter, I am using the framework
of Lambrecht (1994). I will not consider locational or temporal phrases that are not
directly what the comment is about to be topics, as they simply set the scene.
Taking an insight from Shen (2006), I would like to argue here that since
the view of Chinese clause structure as simply topic and comment, with no
grammaticalized categories we might call “subject” or “direct object,” can explain
all of the clause patterns found in Chinese, assuming any other analysis that posits
categories for which there is no need or justification would violate the application
of the principle that theoretical constructs should not be multiplied beyond
necessity (“Occam’s Razor”). In this chapter, I will take a few paragraphs chosen at
random from different sources to show how a simple information structure analysis
can explain all of the structures used, without the need for the grammaticalized
categories of “subject” and “direct object.”
stud of Chinese Book-3.indd 10
2008/11/17 12:51:04 PM
Chinese as a topic-comment language 11
The following paragraph is from the novel Sān Jiā Xiàng, by Ouyang Shan
(1959: 245) (I will first present examples in toto in Chinese characters, then go
through the text line by line):
(1)正說着，门外忽然响起了砰砰砰的急急的敲门声。(2)大家的精神都振作了，神
经也緊張起來了。(3)两个青年男子跳了下地，周炳也唰地一声站了起來。(4)周金
對大家說：(5)“不要慌張。沒有什麼可怕的！什麼時候都不要忘記自己是革命男子
汉！”(6)然后叫周炳去開门，自己站在窗前，仰望着那黑沉沉的天空，慢慢地吸
烟。(7)周炳扭亮了神廳电灯，打開了大门，(8)跳進來一个漂亮而壯健，大眼窩，大
嘴巴的年輕小伙子，(9)原來是楊承輝。(10)他把雨衣一扔，就冲進神樓底，气急敗
坏地說：(11)“坏了，坏了！出事儿了！反革命分子动手了！快走吧，走把，走吧！
(1) 正說着，门外忽然响起了砰砰砰的急急的敲门声。
zhèng shuō-zhe, mén-wài
hūrān
xiǎng-qǐ-le
just talk-DUR door-outside suddenly make sound-INCHO-PFV
pēngpēngpēng de
jí-jí
de
qiāo-mén-shēng.
[ONOMA
ASSOC urgent-urgent ASSOC knock-door-sound]
Just as (he) was talking, suddenly outside there was the sound of urgent knocking on the door.
In this first sentence a character named Zhou Rong is speaking, when suddenly
there is a knock on the door. The way this is expressed is with the verb xiǎng “make
a sound.” When using this verb, the reference to the thing which makes the sound
can appear before or after the verb, depending on whether that thing is topical or not.
In the case of (1), the making of the sound of knocking is presented as an event. As
discussed in LaPolla (1995) and LaPolla and Poa (2005), events are often presented
as thetic statements, that is, statements without topics. To achieve this effect, any
referent mentioned must appear in post-verbal position, to avoid it being interpreted
as a topic, such as the yǔ “rain” in xià yǔ le [fall rain CSM] “It has started raining.”
In (1) the sound of knocking is not presented as a topic, and so appears post-verbally.
In other cases, such as if we were talking about the phone and then wanted to say
that it rang, we would say diànhuà xiǎngle [telephone make.sound-PFV] “the phone
rang,” with the reference to the telephone in topic position. The information structure
principle can explain the reason for the structure of (1) and the difference between (1)
and other possible structures involving the same verb, whereas the assumption of the
grammaticalization of “subject,” topic-prominence, or SVO word order cannot.
(2) 大家的精神都振作了，神经也緊張起來了。
1
dājiā de
jīngshén dōu zhènzuò-le, 2 shénjīng yě jǐnzhāng-qǐlái-le.
everyone ASSOC spirit
all rouse-CSM nerve
alsotighten-INCHO-CSM
Everyone roused up, and started to get nervous.
stud of Chinese Book-3.indd 11
2008/11/17 12:51:04 PM
12 Randy J. LaPolla
In the first clause of (2) we have a topic, dàjiā de jīngshén “everyone’s spirits,”
and the comment dōu zhènzuò-le “all roused.” This clause is interesting in that the
same verb, which is an activity verb, can be used with an actor controlling the spirit
as topic, as in dàjiā yào zhènzuò jīngshén [everyone want rouse spirit] “Everyone
should rouse up,” or dàjiā yào zhènzuò (without the mention of spirit), but in (2)
it would not be possible to add an actor argument, as the actual actor appears as
the possessor of the spirit. This is not a passive clause (Chinese in fact does not
have true passive clauses; see LaPolla 1988b, 1990); it is simply a different topiccomment structure, with the spirit(s) as the topic as opposed to having the actor as
the topic. In the second clause of (2), there is a topic shénjīng “nerves,” and it is
understood as having the same possessor as the spirit mentioned in the first clause
(i.e. “everyone”). This is followed by the comment that they began to tighten up (i.e.
the people began to get nervous—the topic here also could have been “everyone,”
but the author chose to contrast everyone’s spirit and nerves as topics in saying the
people roused up and got nervous). The verb in this clause is a state verb, but the
clause has the sense of an achievement predication with the addition of the change
of state marker, and this allows the parallel with the predication of the first clause
marked by yě “also.”
(3) 两个青年男子跳了下地，周炳也唰地一声站了起來。
1
li ng-ge qīngnián nánzi tiào-le-xià
dì,
[two-CL youth male] jump-PFV-down ground
2
Zhōu.B ng yě shuā-de
yī-shēng zhàn-le-q lái.
PN
also ONOMA-ADV one-sound stand-PFV-INCHO
Two male youths jumped down to the floor, Zhou Bing also stood up with a
swoosh.
In (3) we have a topic in the first clause, two male youths, with the comment
that they jumped down to the floor, and this first topic is contrasted with the second
topic, Zhou Bing, with the comment that he stood up with a swoosh. The use of
yě “also” shows that the author is treating the second comment as parallel with the
first one, even though it is not the same sort of action (both reflect the coming to
attention and nervousness mentioned in the previous two clauses). It will be noticed
that the topic in the first clause is represented as if it is an unidentifiable referent.
Though it is possible to understand this clause assuming the two men mentioned
are unidentifiable, in fact given the larger context of the book we can infer that the
two young men referred two are Zhou Jin and Zhou Rong, who had been said to be
lying on the beds in the room. This sort of reference probably would only be made
in a written genre such as this passage.4
stud of Chinese Book-3.indd 12
2008/11/17 12:51:04 PM
Chinese as a topic-comment language 13
(4) 周金對大家說：“不要慌張。沒有什麼可怕的！什麼時候都不要忘記自己
是革命男子汉！”
Zhōu.Jīn duì
dājiā
shuō:
PN
towards everyone say
1
“Bù-yào
huāngzhāng. 2 Méi-y u
shénme kěpà
de!
NEG-want fluster NEG.PFV-exist what frightening ASSOC
3
Shénme shíhòudōu bù-yào
wàngjì zìj shì
gémìng
nánzihàn!”
what time all NEG-want forget self COPULA revolution male
Zhou Jin said to everyone: “Don’t panic. There is nothing to be afraid of! (You)
should always remember you are revolutionary men!”
In (4) we have a new topic, Zhou Jin, and the comment that he said something
to everyone, and then the quote of what he said. There is no reference to anyone he
is talking to (other than zìjǐ “self”), but we can infer that in the first and third clause
of the quote he is talking to and about everyone. The middle clause is an existential
clause, where the existent appears after the verb, again because it is not topical, but
instead focal. This same sort of clause could instead have a preverbal noun phrase,
if that noun phrase represented a topical referent, e.g. nàxiē huò yǐjīng méi-y u-le
[those goods already NEG.PFV-exist-CSM] “Those goods are gone (no longer
exist).” Again, the information structure principle can explain this difference, and
also show the cognacy of these two clause types, but the assumption of “subject,”
“topic-prominence,” or SVO structure, cannot.
(5) 然后叫周炳去開门，自己站在窗前，仰望着那黑沉沉的天空，慢慢地吸烟。
1
Ránhòu jiào Zhōu.Bīng qù kāi mén,
after.that tell PN
go open door
2
zìjǐ zhàn zài chuāng qián,
self stand LOC window before
3
yǎngwàng-zhe nà hēichénchén de
tiānkōng,
look.up.at-DUR [that very.black
ASSOC sky]
4
mànmàn de
xīyān.
slow
ADV smoke
Then he told Zhou Bing to go open the door, (while) he himself stood in front of
the window, looking at the dark sky, slowing smoking.
In (5) we have a topic chain with four parallel clauses all about the same topic,
Zhou Jin, who had been mentioned in (4), but is simply assumed in the clauses in (5)
(the quoting clause, but not the quoted clauses of (4), is part of the topic chain that
includes the clauses in [5]), and there is the comment in the first clause that he told
stud of Chinese Book-3.indd 13
2008/11/17 12:51:04 PM
14 Randy J. LaPolla
Zhou Bing to open the door, and a comment in the second clause, that he himself
stood in front of the window. In this second clause the word zìjǐ “self” appears in
apposition to the unstated topic, Zhou Jin. In the third and fourth clauses the same
topic is assumed, and two comments are made, that he looked up at the dark sky and
slowly smoked, but the comment in the third clause is understood (because of the
appearance of -zhe, which in this context marks simultaneous action) as modifying
the action of the fourth comment, and so the two clauses together are understood
as “(he) slowly smoked while looking up at the sky.” This is also understood as
happening while he stood in front of the window, so these last three clauses are
semantically more tightly related to each other than any is to the first clause. Notice
that in the third clause we have the opposite situation of what we saw in (3): here
the noun phrase “that black sky” is overtly marked as identifiable, yet appears in
focus position, showing that identifiability and topicality are separate statuses.5
(6) 周炳扭亮了神廳电灯，打開了大门，
1
Zhōu.Bǐngniǔ-liàng-le shén-tīng-diàn-dēng, 2 dǎ-kāi-le dà-mén,
PN
twist-bright-PFVspirit-hall-electric-light hit-open-PFV big-door
Zhou Bing turned up the shrine hall light and opened the main door,
In (6) we have a person who can be understood as an actor as topic of both
clauses, and except for the actor not appearing overtly in the second clause (it is
assumed to be the same as the first, but this is not an obligatory understanding
forced by the grammar the way it is in English cross-clause coreference), they seem
like simple SVO clauses. The question is why these clauses are the way they are.
Is the reference to Zhou Bing before the verb because he is subject, or because he
is topic? Is the reference to the light and the door after the verb because they are
objects or because they are focal? The references to the light and the door could
come before the verb, but then they would be understood as topics, that is, what
the story is about. This part of the story is not about the light and the door, but
about Zhou Bing, and so they appear after the verb. As for the initial noun phrase,
it seems very much that in the structure NP1 V NP2, where one of the noun phrases
can be understood as agent of a transitive clause, that noun phrase must precede the
verb. This is probably the reason why Li and Thompson (1981) still had “subject”
as part of their description of Chinese (but defined in semantic rather than structural
terms), even though they understood the pragmatic influences on word order.
In earlier work (e.g. LaPolla 1990), I also said that aside from the information
structure principle, there was also a semantic rule that agents of transitive clauses
appear before the verb. This was certainly not a syntactic rule, as it is not a matter
of NP1 in an NP1 V NP2 always being an agent, as in English, but only said to
stud of Chinese Book-3.indd 14
2008/11/17 12:51:04 PM
Chinese as a topic-comment language 15
hold in clauses where NP1 can be understood as an agent, and does not include
the agents of intransitive clauses, so even if there were such a rule, it would not be
related to “subject” as a category of all clauses. If it were a syntactic rule related
to “subject,” the way it is in English, NP1 in any NP1 V NP2 structure or NP1 V
(where the verb is not marked as passive) structure would have to be understood as
agent, but this is obviously not the case in Chinese. I later realized this rule does not
exist, as Y. R. Chao had said if there exists N1 V N2, it is not necessarily the case
that N1 is the agent and N2 the patient: “A corollary to the topic-comment nature
of predication is that the direction of action in an action verb in the predicate need
not go outward from subject to object. Even in an N-V-N´ sequence, such as 狗咬
人 gǒu yǎo rén [dog bite man], it is not always certain that the action goes outward
from N to N´” (1968: 70, with pinyin replacing Chao’s transcription). Lü Shuxiang
(1979: 70–72) also argued against any sort of determination of zhǔyǔ (“subject”)
on the basis of semantics, and he gives examples such as that in (11), below, as
examples of an agent appearing post-verbally. Examples of the type 這鍋飯吃了三
个人 zhè guō fàn chī-le sān-ge rén [this pot rice eat three-CL person] “This pot ate
(fed) three people” are also relevant here. Given these examples, we cannot say that
in an NP1 V NP2 structure where one noun phrase could be understood as an agent
NP1 must be understood as the agent, unless, as some have tried with the last type
of example, we say there are really two different verbs involved. For me it is much
simpler to assume the null hypothesis, that there is no difference in the verb, but a
difference in what appears as topic. This was Lü Shuxiang’s position, and it allows
us to explain much more of the grammar of Chinese with a single general principle
than to have ad hoc unmotivated explanations for each structure.
(7) 跳進來一个漂亮而壯健、大眼窩、大嘴巴的年輕小伙子，原來是楊承輝。
1
tiào-jìn-lái
yī-ge
piàoliàng ěr zhuāngjiàn,
jump-enter-come [one-CL beautiful and strapping
dà yǎnwō,
dà zǔiba de
niánqīng xiǎohuǒzi
big eye.socket big mouth ASSOC young
fellow]
2
yuánlái shì
Yáng.Chénghuī.
originally COPULA PN
and a good-looking and strapping young fellow with big eyes and a big mouth
jumped into (the room); it was Yang Chenghui.
In the first clause in (7) we have the same verb as in (3), tiaò “jump,” but
unlike in (3), the reference to the one who jumped does not appear before the verb,
but after the verb. This is because the utterance is not about the young man that
jumped in, i.e. he is not the topic of the utterance, but is presenting an event, the
stud of Chinese Book-3.indd 15
2008/11/17 12:51:04 PM
16 Randy J. LaPolla
entrance of this young man. At the same time it is introducing the young man into
the discourse, and then he becomes the topic of the following clause, where his
name is given.
(8) 他把雨衣一扔，就冲進神樓底，气急敗坏地說：“坏了，坏了！出事儿了！
反革命分子动手了！快走吧，走把，走吧！
1
Tā bǎ yǔyī
yī rēng,
3sg BA raincoat one throw
2
jiù chōng-jìn shén-lóu-dǐ,
then rush-enter spirit-house-bottom
3
qìjíbàihuài
de shuō:
flustered.and.exasperated ADV say
1
3
“Huài-le, 2 huài-le!
Chū shìr le!
bad-CSM bad-CSM exit affair CSM
4
Fǎngémìngfènzi dòng shoǔ le!
reactionaries
move hand CSM
5
6
7
Kuài zǒu bā,
zǒu bā,
zǒu bā!”
fast leave HORT leave HORT leave HORT
As soon as he threw aside his raincoat he rushed into the room behind the shrine
hall, and in a very flustered and exasperated way said, “Something’s gone terribly
wrong! Something’s gone terribly wrong! Something bad’s happened! The
reactionaries have taken action! Leave quickly, leave, leave!”
The same referent (the young man) is assumed as the topic of the first three
clauses of (8). This is a new topic chain, and so there is reference to him in the first
clause with a third person singular pronoun, and the same referent is understood as
the topic of the next two clauses (again, not obligatorily, but inferred from context),
the last of which introduces a quote. In the first clause of (8) there is a new referent,
the raincoat, but it is not introduced and treated as a referent of concern; it appears
as if it was already in the scene, as a secondary topic following the auxiliary verb
bǎ, which gives a sense of disposal of the raincoat. This requires some pragmatic
accommodation on the part of the reader, relying on background assumptions that
it is raining and so anyone coming in would be wearing a raincoat and need to take
it off when they come in. The reference to the raincoat appears in this position in
the clause because the action is what is in focus, and so the verb appears in final
position (the first, second, and third clauses of [8] are a series of actions by the
young man), and mentioning the throwing of the raincoat adds to the sense of
urgency of his actions. In other contexts the reference to the raincoat could appear
in post-verbal position, but if that structure were used here there would be more
stud of Chinese Book-3.indd 16
2008/11/17 12:51:04 PM
Chinese as a topic-comment language 17
a sense that the raincoat is what is in focus, and we would expect the following
clauses to say something about what happened to the raincoat or as a result of
throwing the raincoat. In the second clause we again have an action like in the first
clause of (7), rushing into a room, and by the same referent, but in this case the
rushing into the room is not presented as an event, but as one of a series of actions
by the same topic, and so if there were representation of the referent, it would
appear in topic position.
In the quoted part of (8), there is no reference to the topic of the first two
clauses, but we can understand that it is the general situation that is the topic. In
the third clause of the quote, there is no topic; again an event is presented, that
something bad happened. The fourth clause of the quote explains what that event
was, taking the reactionaries as topic, and saying that they have started to do
something (arrest people). Having the reactionaries as topic with the predication 动
手了 dòngshǒu-le “took action” gives the impression the men had been anticipating
such an action by the reactionaries (we also know this from the larger context). The
last few clauses of the quote are imperatives, urging everyone to run away.
Consider also example (9) (from blog.ytcnc.net/user1/abc0805/
archives/2006/2275.html):
(9) 特別喜歡下雪。喜歡下雪后白茫茫的一片很純淨的感覺。希望今年的第一場
雪早点下吧！
1
Tèbié
xǐhuān xià xuě.
especially like fall snow
2
Xǐhuān xià xuě hòu báimángmáng
de
like
fall snow after glisteningly.white ASSOC
yī-piàn hěn chúnjìng de
gǎnjué.
one-CL very pure
1
ASSOC feeling
Xīwàng jīnnián de
dì-yī-chǎng xuě zǎo diǎn xià bā!
hope this.year ASSOC ORD-one-CL snow early a.bit fall HORT
(I) especially like snow (lit: “the falling of snow”). (I) like the feeling of purity of
the glisteningly white snow after a snowfall. (I) hope this year the first snow (of
the season) will fall a bit earlier!
In this example the same noun, xuě “snow,” appears after the verb in the first two
clauses, but before the verb in the last clause. This is because in the first two clauses
“the falling of snow” is presented as an event, and so the reference to snow must
follow the verb in order not to be interpreted as a topic, while in the last clause the
snow is a topic, and the comment about this topic is that it falls early (in the year).
stud of Chinese Book-3.indd 17
2008/11/17 12:51:04 PM
18 Randy J. LaPolla
As discussed at length by Lü Shuxiang (1979: 72–73), in many cases where
a clause has representations of two referents, the two representations can appear
in either order, with no difference in the interpretation of semantic or grammatical
roles. The only difference is which of the two referents is topical and which is focal.
Professor Lü gave examples such as those in (10) and (11):
(10) a.窗户已经糊了紙
chuānghu yǐjīng hú-le
zhǐ
window already paste-PFV paper
The window has already been pasted with paper
b.紙已经糊了窗户
zhǐ
yǐjīng hú-le
chuānghu
paper already paste-PFV window
The paper has already been pasted on the window
(11) a.這个人沒有騎過马
zhè-ge rén
méi-yǒu
qǐ-guò
mǎ
this-CL person NEG.PFV-exist ride-EXP horse
This person has never ridden a horse
b.這匹马沒有騎過人
zhè-pǐ mǎ méi-yǒu
qǐ-guò rén
this-CL horse NEG.PFV-exist ride-EXP person
This horse has never (been) ridden (by) a person
Examine the following natural example (from bulo.cn.yahoo.com/blog/blog_article.
php?bname=hungtoyeung&mid=347):
(12) 雖然一间屋住了十多个人, (13) 一層樓里大铁门內有大約12个屋, (14) 近100
人吃喝拉撒，一年到头拥擠在只有約800平方米的一个空间里, (15) 但他們
的情緒从表面上看是很平靜和安詳的，⋯⋯
(12)
雖然一间屋住了十多个人,
suīrán yī-jiān wū
zhù-le
shí-duō-ge rén
although one-CL room live-PFV ten-more-CL person
Although one room housed more than ten people,
In this example, the initial topic, mentioned in (12), is a room, and the
comment about this room is that more than ten people live there. The room is the
stud of Chinese Book-3.indd 18
2008/11/17 12:51:04 PM
Chinese as a topic-comment language 19
topic and the people who live there are within the focus. Although generally topics
are specific and identifiable, in this context, the room talked about is not specific,
but “a room” of the building in the generic sense. Notice that in Chinese the same
verb zhù “live” can be used whether the topic is the place lived or the people living
in the place (see [16] below), whereas in English two different verbs, live vs. house,
must be used to achieve the same effect. This difference is also due to the difference
of topic-comment structure vs. subject-predicate structure.
(13) 一層樓里大铁门內有大約12个屋,
yī-céng lóu
lǐ
dà tiě-mén nèi
yǒu dàyuē 12 ge wū
one-floor building within big iron-door inside exist about 12 CL room
On one floor inside the big iron door were about twelve rooms.
The clause in (13) is an existential clause, stating that on each floor there are
about twelve rooms. Here the floor is a locus rather than a topic (it has a scenesetting function, i.e. identification of a time or place).6
(14) 近100人吃喝拉撒，一年到头拥擠在只有約800平方米的一个空间里,
1
jìn
100 rén
chī-hē-lā-sā,
close.to 100 person eat-drink-shit-piss
2
yī-nián dào-tóu
yōngj
one-year arrive-end crowded
zài zhǐ yǒu yuē 800 píngfāng mǐ
de
yī-ge kōngjiān lǐ,
LOConlyexist about 800 square meters ASSOC one-CL space
within
close to 100 people doing all of their activities all year long crowded into a space
of only 800 square meters,
The two clauses in (14) express the logical consequence of ten people per room
and twelve rooms on one floor: that close to 100 people carry out all their daily
activities within a mere 800 square meters of space. In these clauses the people
have become the topic, rather than the space, which is now within the focus. All
four of these clauses are within the scope of the initial suīrán “although.”
(15) 但他們的情緒从表面上看是很平靜和安詳的，⋯⋯
dàn tāmen de
qíngxù cóng biǎomiàn shàng kàn
but 3pl ASSOC mood from surface on
look
shì
hěn píngjìng hé ānxiáng de, . . .
COPULA very peaceful and serene NOM
but their mood was seemingly peaceful and serene.
stud of Chinese Book-3.indd 19
2008/11/17 12:51:04 PM
20 Randy J. LaPolla
The last clause, which takes the mood of these people as the topic, is the
contrasting statement, that their mood is seemingly peaceful and contented.
We can also contrast the structure of (12) with the following example, the
title of a news item on an on-line news service (www.epochtimes.com/b5/5/3/23/
n862553.htm, February 19, 2007), where the order of the two noun phrases is the
reverse of that in (12):
(16) 中国女工毛里求斯遭遇：近千人住一间屋
1
Zhōngguó nyŭ-gōng
Máolı̌ qiúsı̄ zāo-yù:
Chinese woman-worker Mauritius suffer-meet
2
jı̀n
qiān
rén
zhù yī-jiān wū.
close.to thousand people live one-CL room
Chinese female worker suffers in Mauritius: close to a thousand people live in
one room.
Here we see that the people are taken as the topic rather than the room, but
other than that the semantics are the same.
We can see from these short passages that there are many instances where the
reference to some referent can either precede or follow the verb, and the deciding
factor is not whether the referent is “subject” or “direct object,” and not whether it
is identifiable (“definite”) or unidentifiable (“indefinite”), but whether it is topical
or focal.
This sort of analysis can in fact explain all of the other types of “odd” clauses
in Chinese, such as the double topic structure and the “split referent” structure.
Another type it can explain easily is the type as in (17) (from Lián Chéng Jué (連
城訣), by Jin Yong (金庸), Chapter 6; http://louisville.edu/journal/weiming/wuxia/
lian06.txt):
(17)她死了一匹马, 便這么哭个不住。
1
2
tā sǐ-le
yī-pǐ
mǎ,
biàn zhème
kū ge bù-zhù.
3sg die-PFV one-CL horse then this.much cry CL NEG-stop
She had a horse die on her, and she cries this much without stopping.
If we assume an SVO or any other “subject”-based analysis of Chinese, we run
into serious problems with the type of structure in the first clause of (17), which
is quite common in Chinese. If we try to say that “she” is the “subject” and “one
horse” is the object, then we must assume that “die” is a transitive verb, or at least
has a transitive use. But this goes against what Chinese speakers feel about this
sentence. It is not that “she” caused the death of the horse, but that the horse died,
stud of Chinese Book-3.indd 20
2008/11/17 12:51:05 PM
Chinese as a topic-comment language 21
and this has affected her in some way. If we use the topic comment analysis, we can
see that “she” is the topic, and “die a horse” is presented as an event. It is the horse
that died, but the dying is not presented as a comment about the horse, that it died,
but as an event of horse dying, and this event functions as the comment about “she.”
In most discussions of Chinese, the double topic construction, as in (18), and
the “split referent” construction, as in (19), are not discussed together, and it is
not felt that they are related, but they are in fact explained by the same principle,
that topical elements come before the verb and focal elements come after the verb.
(Example [18] is from an online diary: http://spaces.huash.com/?111533/action_
viewspace_itemid_210107.html, posted January 25, 2007, 22:44:40; (19) is from
http://book.msn.com.cn/n/a/34198/326436.shtml, both accessed February 23,
2007.)
(18)丫头你提了个問題很嚴重，睡覺前我要想一下，想不出就不睡了,
但是現在我肚子餓，所以我要先吃点东西。
Yātou nǐ
tí-le
ge wèntí
hěn yánzhòng, . . .
girl 2sg raise-PFV CL question very serious
1
2
3
dànshì xiànzài wǒ dùzi è,
suǒyǐ wǒ yào xiān chī dōngxi.
but
now
so
1sg want first eat thing
1sg belly hungry
Girl, you’ve raised a question that is very serious, . . . but now I am (my belly is)
hungry, and so I want to eat something first.
(19)等他那几个小菜做好的時候，我已经飢腸轆轆了。我埋怨他动作太慢，害我
餓了肚子，他却好脾气的說，慢工出細活嘛！
. . . Wǒ máiyuàn tā dòngzuò tài màn, hài wǒ è-le
dùzi, . . .
1sg complain 3sg movement too slow harm 1sg hungry-CSM belly
. . . I complained that his movements were too slow, (and) caused me to get
hungry, . . .
In these two examples, though both involve the expression for saying one
is hungry, which involves reference to one’s belly, there is a difference in the
placement of the reference to “belly.” In (18) it appears in preverbal position as
secondary topic in a double-topic structure of the type [topic [topic-comment]comment].
In (19) it appears in post-verbal position. This difference is due to a difference in
information structure. In both cases “I” am the main topic about which a comment
is being made, and in both cases the other referent is the same, “(my) belly,”7 but in
(18) the comment about “I” includes a secondary topic, as “I” am saying something
about “(my) belly,” that it is hungry, and this statement about “(my) belly”
constitutes the comment about “I.” In (19) there is no secondary topic; “I” am not
stud of Chinese Book-3.indd 21
2008/11/17 12:51:05 PM
22 Randy J. LaPolla
saying something about “(my) belly.” The referent “(my) belly” is still involved
in the comment made about “I,” but in this case the comment takes the form of an
event, getting hungry, and as “(my) belly” is not topical, it appears in post-verbal
position.
Notice also that in the first clause of (18), rather than saying tí-le ge yánzhòng
de wèntí [raise-PFV CL serious ASSOC question] “raised a serious question,”
with “serious” in a relative clause structure modifying “question,” the author has
“serious” as a second predication. This can also be explained with an information
structure account. In the relative clause structure, “serious” is treated as part of the
same focus as “raise a question,” whereas in (18) “raise a question” is presented in
one focus, and the fact that the question is serious is presented as a separate focus,
a separate comment on the referent just introduced. That is, it is equivalent to two
clauses, where the first introduces a referent and the second makes a comment
on it (“raised a question” + “the question is serious”), but it is collapsed into a
single structure with two foci (see LaPolla 1995). This sort of construction is most
common with an existential verb as the first verb (e.g. wǒ yǒu ge péngyǒu chū-le
chēhuò [1sg exist CL friend happen-CSM car.accident] “I have a friend who had a
car accident”—from www.pcauto.com.cn/playcar/owner_report/rcgs/0410/153172/
html).
In this chapter I hope to have shown that an information structure analysis
can elegantly explain all of the clause patterns found in these Chinese passages,
including many that are problematic for other analyses. As that is the case, there is
no need to posit any grammaticalized categories, such as “subject,” to explain the
structure of the clause in Chinese. In fact assumption of “SVO”8 structure would be
problematic given the clause patterns we find in Chinese.
stud of Chinese Book-3.indd 22
2008/11/17 12:51:05 PM